U.S. patent number 7,305,967 [Application Number 11/705,427] was granted by the patent office on 2007-12-11 for control apparatus for an internal combustion engine.
This patent grant is currently assigned to Mitsubishi Electric Corporation. Invention is credited to Hideki Hagari, Hideki Nishimura.
United States Patent |
7,305,967 |
Hagari , et al. |
December 11, 2007 |
Control apparatus for an internal combustion engine
Abstract
In throttle control, a throttle opening is set with sufficient
control accuracy in accordance with the operating state of an
engine despite variations in a throttle body and various kinds of
sensors. A target effective opening area is calculated from a
target amount of intake air, an atmospheric pressure, an intake
pipe internal pressure and an intake air temperature by using a
flow rate formula for a throttle type flow meter. A target throttle
opening is calculated from a correlation map. An actual effective
opening area is calculated from the amount of intake air, the
atmospheric pressure, the intake pipe internal pressure, and the
intake air temperature by using the above-mentioned flow rate
formula, and a learning throttle opening is calculated from the
correlation map. The target throttle opening is corrected by a
throttle opening learning value calculated from a deviation between
the target throttle opening and the learning throttle opening.
Inventors: |
Hagari; Hideki (Tokyo,
JP), Nishimura; Hideki (Tokyo, JP) |
Assignee: |
Mitsubishi Electric Corporation
(Tokyo, JP)
|
Family
ID: |
38792757 |
Appl.
No.: |
11/705,427 |
Filed: |
February 13, 2007 |
Foreign Application Priority Data
|
|
|
|
|
Aug 29, 2006 [JP] |
|
|
2006-231950 |
|
Current U.S.
Class: |
123/403; 701/103;
123/480; 123/478 |
Current CPC
Class: |
F02M
35/10295 (20130101); F02M 35/1038 (20130101); F02D
9/02 (20130101); F02D 41/18 (20130101); F02D
11/105 (20130101); F02M 35/10386 (20130101); F02D
41/2464 (20130101); F02D 41/0002 (20130101); F02M
35/10052 (20130101); Y02T 10/42 (20130101); F02D
2200/703 (20130101); Y02T 10/40 (20130101); F02D
2200/0406 (20130101); F02D 41/2448 (20130101); F02D
2200/0414 (20130101); F02D 2200/0404 (20130101) |
Current International
Class: |
F02D
9/08 (20060101); B60T 7/12 (20060101) |
Field of
Search: |
;123/402,403,478,480,462
;701/103 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Primary Examiner: Kwon; John T.
Attorney, Agent or Firm: Sughrue Mion Pllc.
Claims
What is claimed is:
1. A control apparatus for an internal combustion engine
comprising: a throttle valve that is arranged in an intake passage
of said internal combustion engine; a throttle opening control
section that variably controls an amount of intake air supplied to
said internal combustion engine by changing an effective opening
area of said intake passage thereby to control a throttle opening
of said throttle valve; an operating state detection section that
detects an operating state of said internal combustion engine, and
includes an intake air amount detection section that detects the
amount of intake air supplied to said internal combustion engine,
an atmospheric pressure detection section that detects pressure at
an atmospheric side of said throttle valve as an atmospheric
pressure, an intake pipe internal pressure detection section that
detects pressure at said internal combustion engine side of said
throttle valve as an intake pipe internal pressure, and an intake
air temperature detection section that detects an intake air
temperature at an atmospheric side of said throttle valve; a target
intake air amount calculation section that calculates a target
amount of intake air based on the operating state of said internal
combustion engine; a target effective opening area calculation
section that calculates a target effective opening area of said
throttle opening control section by applying said target amount of
intake air, said atmospheric pressure, said intake pipe internal
pressure and said intake air temperature to a flow rate formula for
a throttle type flow meter; a target throttle opening calculation
section that calculates a target throttle opening from said target
effective opening area by using a correlation map between the
effective opening area and the throttle opening, which are suited
to each other in advance, of said throttle opening control section;
an actual effective opening area calculation section that
calculates an actual effective opening area of said throttle
opening control section by applying said amount of intake air, said
atmospheric pressure, said intake pipe internal pressure and said
intake air temperature to said flow rate formula for a throttle
type flow meter; and a learning throttle opening calculation
section that calculates a learning throttle opening from said
actual effective opening area by using said correlation map;
wherein said throttle opening control section includes a throttle
opening learning value calculation section that calculates a
throttle opening learning value based on a deviation between said
target throttle opening and said learning throttle opening; and
said throttle opening control section controls said throttle
opening based on a learning corrected target throttle opening which
is obtained by correcting said target throttle opening by said
throttle opening learning value.
2. The control apparatus for an internal combustion engine as set
forth in claim 1, wherein said throttle opening learning value is
distributed to and stored in at least one of a real-time learning
value being updated in real time and a long time learning value
corresponding to each learning region according to an effective
opening area axis of said correlation map.
3. The control apparatus for an internal combustion engine as set
forth in claim 1, wherein said long time learning value is stored
in at least one of a first learning region corresponding to said
target effective opening area and a second learning region
corresponding to said actual effective opening area.
4. The control apparatus for an internal combustion engine as set
forth in claim 1, wherein said long time learning value is limited
in such a manner that said correlation map and a relation between
an effective opening area of said throttle opening control section
to which said long time learning value is added and said throttle
opening become monotonously increasing.
5. The control apparatus for an internal combustion engine as set
forth in claim 1, wherein the updates of said real-time learning
value and said long time learning value is inhibited when a
pressure ratio of said intake pipe internal pressure to said
atmospheric pressure indicates a value equal to or larger than a
first predetermined value.
6. The control apparatus for an internal combustion engine as set
forth in claim 1, wherein in a period in which the time elapsed
after a time change rate of said target amount of intake air has
reached a second predetermined value or more indicates a value less
than or equal to a third predetermined value, said real-time
learning value is reset, and the update of said long time learning
value is inhibited.
7. The control apparatus for an internal combustion engine as set
forth in claim 1, wherein said throttle opening control section
includes a backup memory; when said engine is stopped or when a
power supply for said control apparatus is turned off, said
real-time learning value is reset, and said long time learning
value is held in said backup memory; in a period of time in which
the time elapsed after the starting of said engine indicates a
value within a fourth predetermined value, the update of said
real-time learning value is inhibited; in a period of time in which
the time elapsed after the starting of said engine indicates a
value that is equal to or larger than said fourth predetermined
value and within a fifth predetermined value, the update of said
long time learning value is inhibited; and when the number of
revolutions per minute of said internal combustion engine indicates
a value less than or equal to a sixth predetermined value that is
lower than a target number of revolutions per minute of said engine
during idling, the update of said long time learning value is
inhibited.
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a control apparatus for an
internal combustion engine capable of controlling the opening of a
throttle valve so as to obtain a target amount of intake air.
2. Description of the Related Art
In recent years, there has been proposed a control apparatus for an
internal combustion engine that can obtain excellent driving
performance by using, as a requested value of a driving force from
a driver or a vehicle side, the output shaft torque of the internal
combustion engine (hereinafter also referred to simply as an
engine) which is a physical quantity directly acting on the control
of the vehicle, and by deciding, as an output target value of the
engine in the form of the output shaft torque, engine control
quantities in the form of an amount of air, an amount of fuel and
ignition timing.
In addition, it is generally known that a control quantity having
the greatest influence on the engine output shaft torque among the
engine control quantities is the amount of air, and there has also
been proposed a control apparatus for an internal combustion engine
that is capable of controlling the amount of air with a high degree
of precision (see, for example, a first patent document: Japanese
patent application laid-open No. H11-229904).
In the above-mentioned first patent document, in the control
apparatus for an internal combustion engine that controls the
opening of a throttle valve by driving an actuator provided in
association with the throttle valve, a target throttle effective
opening area is calculated by applying a target amount of intake
air corresponding to a target engine torque to an orifice flow rate
expression which is based on a differential pressure across the
throttle valve, an air passing area, and a specific throttle
opening is set so as to attain the target throttle effective
opening area thus calculated.
Thus, when the throttle opening that attains the target amount of
intake air is calculated by applying it to the orifice flow rate
expression, the target amount of intake air can be adequately
attained even in cases where an environmental condition such as an
atmospheric pressure, an intake air temperature, etc., has changed,
or where exhaust gas recirculation (hereinafter referred to as
"EGR") for introduction of an exhaust gas into an intake pipe is
carried out.
In the conventional control apparatus for an internal combustion
engine, for example in case of the first patent document, in a
throttle valve of which an effective opening area changes in
accordance with the operating state of the engine, a flow
coefficient, which greatly influences the shape and the opening
area of the throttle valve, is obtained from the number of
revolutions per minute of the engine and the pressure ratio of an
intake pipe internal pressure (hereinafter referred to as an
"intake manifold pressure") and the atmospheric pressure. As a
result, it is difficult to accurately set the flow coefficient in a
state in which the degree of opening and an effective opening area
of the throttle valve are not decided.
Accordingly, there is the following problem. That is, the target
throttle effective opening area to obtain the target amount of
intake air can not be calculated accurately, so there arises a
deviation between the target amount of intake air and the actual
amount of intake air, and besides a lot of labor is required to
obtain the flow coefficient and to set a map therefor.
In view of the above, it can be considered that the target throttle
effective opening area in the form of the product of the throttle
effective opening area and the flow coefficient is calculated by
applying the target amount of intake air to the orifice flow rate
expression which is based on the differential pressure across the
throttle valve, the air passing area, and the flow coefficient, and
the target throttle opening is calculated by using the relation
between the effective opening area and the throttle opening which
are suited to each other in advance, so that a throttle opening to
obtain the target amount of intake air is thereby calculated
without setting the flow coefficient. In this case, however, there
is the following problem. That is, even with the same throttle
opening, there will arise a variation in the actual opening area
and/or the flow coefficient resulting from the manufacturing
variation of individual throttle bodies, etc., so the amount of
intake air changes depending upon the individual throttle
bodies.
Further, there takes place a variation in the calculated effective
opening area due to the variation and/or estimation error of
various kinds of sensors that measure the intake manifold pressure,
the atmospheric pressure, the intake air temperature, etc. As a
result, there is a problem that there arises a variation in the
actual amount of intake air with respect to the target amount of
intake air due to the variation of the throttle body and the
various kinds of sensors, various kinds of estimation errors,
etc.
SUMMARY OF THE INVENTION
Accordingly, the present invention is intended to obviate the
problems as referred to above, and has for its object to obtain a
control apparatus for an internal combustion engine which can learn
and correct the relation between an effective opening area and a
throttle opening in such a manner that upon calculation of a
throttle opening for obtaining a target amount of intake air, the
target amount of intake air can be adequately attained with respect
to variations in the throttle body and various sensors or various
kinds of estimation errors.
Bearing the above object in mind, a control apparatus for an
internal combustion engine according to the present invention
includes: a throttle valve that is arranged in an intake passage of
the internal combustion engine; a throttle opening control section
that variably controls an amount of intake air supplied to the
internal combustion engine by changing an effective opening area of
the intake passage thereby to control a throttle opening of the
throttle valve; and an operating state detection section that
detects an operating state of the internal combustion engine, and
includes an intake air amount detection section that detects the
amount of intake air supplied to the internal combustion engine, an
atmospheric pressure detection section that detects pressure at an
atmospheric side of the throttle valve as an atmospheric pressure,
an intake pipe internal pressure detection section that detects
pressure at the internal combustion engine side of the throttle
valve as an intake pipe internal pressure, and an intake air
temperature detection section that detects an intake air
temperature at an atmospheric side of the throttle valve. The
control apparatus further includes: a target intake air amount
calculation section that calculates a target amount of intake air
based on the operating state of the internal combustion engine; a
target effective opening area calculation section that calculates a
target effective opening area of the throttle opening control
section by applying the target amount of intake air, the
atmospheric pressure, the intake pipe internal pressure and the
intake air temperature to a flow rate formula for a throttle type
flow meter; a target throttle opening calculation section that
calculates a target throttle opening from the target effective
opening area by using a correlation map between the effective
opening area and the throttle opening, which are suited to each
other in advance, of the throttle opening control section; an
actual effective opening area calculation section that calculates
an actual effective opening area of the throttle opening control
section by applying the amount of intake air, the atmospheric
pressure, the intake pipe internal pressure and the intake air
temperature to the flow rate formula for a throttle type flow
meter; and a learning throttle opening calculation section that
calculates a learning throttle opening from the actual effective
opening area by using the correlation map. The throttle opening
control section includes a throttle opening learning value
calculation section that calculates a throttle opening learning
value based on a deviation between the target throttle opening and
the learning throttle opening, and the throttle opening control
section controls the throttle opening based on a learning corrected
target throttle opening which is obtained by correcting the target
throttle opening by the throttle opening learning value.
According to the present invention, a deviation in the relation
between the effective opening area and the throttle opening is
learned and corrected on the basis of a deviation between the
target amount of intake air and the actual amount of intake air, so
even in case where there are variations in the throttle body and
the various kinds of sensors or various estimation errors, the
throttle opening can be controlled so as to make the amount of
intake air coincide with the target amount of intake air in an
accurate manner.
The above and other objects, features and advantages of the present
invention will become more readily apparent to those skilled in the
art from the following detailed description of preferred
embodiments of the present invention taken in conjunction with the
accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a construction view conceptually showing a control
apparatus for an internal combustion engine according to a first
embodiment of the present invention.
FIG. 2 is a block diagram showing the schematic configuration of
the control apparatus for an internal combustion engine according
to the first embodiment of the present invention.
FIG. 3 is a functional block diagram showing a part of a throttle
opening control section according to the first embodiment of the
present invention.
FIG. 4 is an explanatory view schematically illustrating the
calculation processing of a throttle opening learning value
according to the first embodiment of the present invention.
FIG. 5 is a functional block diagram schematically illustrating a
throttle opening learning value calculation processing part of the
throttle opening control section according to the first embodiment
of the present invention.
FIG. 6 is an explanatory view illustrating the relation of
individual patterns between a CAt-TP map applied according to a
second embodiment of the present invention and an actual
CAt-TP.
FIG. 7 is a functional block diagram schematically illustrating a
storage processing section for a long time learning value according
to the second embodiment of the present invention.
FIG. 8 is an explanatory view schematically illustrating storage
processing for the long time learning value according to the second
embodiment of the present invention.
FIG. 9 is an explanatory view schematically illustrating
monotonously increasing processing for the long time learning value
according to the second embodiment of the present invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
Now, preferred embodiments of the present invention will be
described below in detail while referring to the accompanying
drawings.
Embodiment 1
Referring to the accompanying drawings and first to FIG. 1, therein
is conceptually shown the construction of a control apparatus for
an internal combustion engine according to a first embodiment of
the present invention, and FIG. 2 is a block diagram that shows the
schematic configuration of an engine control part of the control
apparatus for an internal combustion engine according to the first
embodiment of the present invention.
In FIG. 1, at an upstream side of an intake passage that
constitutes an intake system of an engine 1, there are arranged an
air flow sensor 2 that measures the flow rate of intake air Qa
sucked to the engine 1 (hereinafter referred to as "the amount of
intake air"), and an intake air temperature sensor 3 that measures
the temperature of intake air To (hereinafter referred to as "the
intake air temperature").
Here, note that the intake air temperature sensor 3 may be formed
integrally with the air flow sensor 2, or may be formed separately
from the air flow sensor 2. In addition, an element for estimating
the intake air temperature To from other sensor information may be
used in place of the intake air temperature sensor 3 that directly
measures the intake air temperature To.
In the intake system of the engine 1, at an engine 1 side
downstream of the air flow sensor 2, there is arranged a throttle
valve 4 that is adapted to be controlled to open and close for
adjusting the amount of intake air Qa.
A throttle position sensor 5 for measuring the opening degree TP of
the throttle valve 4 (hereinafter referred to as a throttle opening
TP) is attached to the throttle valve 4.
Also, at the engine 1 side downstream of the throttle valve 4,
there are arranged a surge tank 6 that serves to make uniform the
pressure in an intake pipe, and an intake manifold pressure sensor
7 that measures the pressure in the surge tank 6 as an intake pipe
internal pressure (intake manifold pressure) Pe.
Further, connected to the surge tank 6 is an EGR valve 8 that
serves to open and close an EGR tube which is placed in
communication with an exhaust pipe of the engine 1.
Here, note that in place of the intake manifold pressure sensor 7
that directly measures the intake manifold pressure Pe, there any
be used an element for estimating the intake manifold pressure Pe
from other sensor information.
The amount of intake air Qa from the air flow sensor 2, the intake
air temperature To (the temperature at an atmospheric side of the
throttle valve 4) from the intake air temperature sensor 3, the
throttle opening TP from the throttle position sensor 5, and the
intake manifold pressure Pe from the intake manifold pressure
sensor 7 are input to an electronic control unit (hereinafter
referred to as an "ECU") 9 as information indicating the operating
state of the engine 1 together with detection signals from other
unillustrated sensors.
The ECU 9 controls the throttle opening TP of the throttle valve 4
in accordance with the result of calculation based on the operating
state thereby to adjust the amount of intake air Qa, and also
controls to drive a fuel injection system and an ignition system
(not shown) of the engine 1 at required timing, and to open and
close the EGR valve 8 thereby to improve the combustion state of
the engine 1.
In FIG. 2, connected to the ECU 9 are various kinds of sensors 30
which includes, in addition to the above-mentioned group of sensors
2, 3, 5, 7, an atmospheric pressure sensor 10, etc., that detects
the pressure at an atmospheric side of the throttle valve 4 as an
atmospheric pressure Po.
The ECU 9 is provided with an input interface (hereinafter referred
to as an "input I/F") 9a, a processing unit 9b, and an output
interface (hereinafter referred to as an "output I/F") 9c.
The input I/F 9a takes in the detected information from the
above-mentioned sensor group 2, 3, 5, 7, the atmospheric pressure
Po measured by the atmospheric pressure sensor 10, and detection
signals from the other sensors that are included in the various
kinds of sensors 30, and inputs them to the processing unit 9b.
Here, note that in place of the atmospheric pressure sensor 10 that
directly measures the atmospheric pressure Po, there may be used an
element for estimating the atmospheric pressure Po from other
sensor information.
The processing unit 9b in the ECU 9 includes a throttle opening
control section which variably controls the amount of intake air Qa
to be supplied to the engine 1 by controlling the throttle opening
TP of the throttle valve 4 thereby change the effective opening
area of the intake passage.
To this end, first of all, the processing unit 9b calculates a
target torque of the engine 1 based on the input various data
(engine operating state), and then calculates a target amount of
intake air Qa* to achieve the target torque thus calculated.
Subsequently, the processing unit 9b calculates a target effective
opening area CAt* to achieve the target amount of intake air Qa*,
and also calculates a target throttle opening (hereinafter referred
to as a "target opening") TP* to achieve the target effective
opening area CAt*.
Further, the processing unit 9b calculates a control command value
for the EGR valve 8, and also calculates control command values for
other actuators (e.g., injectors of the fuel injection system
arranged in combustion chambers of the engine 1, ignition coils of
the ignition system, etc.) that are included in various kinds of
actuators 40.
Finally, the output I/F 9c in the ECU 9 outputs driving control
signals based on the calculation results of the ECU 9 to the
various kinds of actuators 40 including the throttle valve 4 and
the EGR valve 8.
As a result, the throttle valve 4 is controlled in such a manner
that the throttle opening TP is made to coincide with the target
opening TP*.
Next, reference will be made to the calculation processing, i.e.,
calculation of the target opening TP* to achieve the target amount
of intake air Qa*, executed by the calculation processing part 9b
in the ECU 9 including the throttle opening control section while
referring to a functional block diagram in FIG. 3.
In FIG. 3, the processing unit 9b in the ECU 9 is provided with a
target intake air amount calculation section 90, a target effective
opening area calculation section 11, a sound speed calculation
section 12, a pressure ratio calculation section 13, a
dimensionless flow rate calculation section 14, and a target
opening calculation section 15.
The target amount of intake air calculation section 90 calculates
the target amount of intake air Qa* to achieve the target torque
corresponding to the operating state of the engine 1, and inputs
the calculated value of the target amount of intake air Qa* to the
target effective opening area calculation section 11.
The sound speed calculation section 12 calculates the speed of
sound a.sub.0 in the atmosphere on the basis of the intake air
temperature To, and inputs it to the target effective opening area
calculation section 11.
The pressure ratio calculation section 13 is in the form of a
divider that calculates a pressure ratio Pe/Po of the intake
manifold pressure Pe relative to the atmospheric pressure Po, and
inputs the thus calculated value of the pressure ratio Pe/Po to the
dimensionless flow calculation section 14.
The dimensionless flow rate calculation section 14 calculates a
dimensionless flow rate .sigma. on the basis of the pressure ratio
Pe/Po, and inputs it to the target effective opening area
calculation section 11.
The target effective opening area calculation section 11 calculates
the target effective opening area CAt* of the throttle valve 4
based on the target amount of intake air Qa*, the speed of sound
a.sub.0 and the dimensionless flow rate .sigma. as input
information, and inputs it to the target opening calculation
section 15.
The target opening calculation section 15 calculates the target
opening TP* corresponding to the target effective opening area CAt*
by using a correspondence or correlation map between the effective
opening area CAt and the throttle opening TP that are suited to
each other in advance (a "CAt-TP map" to be described later).
The calculated value of the target opening TP* is input to a
learning basic value calculation section 20 and a learning
corrected target throttle opening calculation section 23 (to be
described later).
Next, reference will be made to the specific calculation processing
functions of the individual calculation sections 11 through 15 in
FIG. 3.
In general, a volumetric flow formula for a throttle type flow
meter is represented by the following expression (1) by using the
amount of intake air Qa (volumetric flow), the speed of sound
a.sub.0 in the atmosphere, the flow coefficient C, and the opening
area At of the throttle valve 4, the intake manifold pressure Pe,
the atmospheric pressure Po, and the ratio of specific heats k.
.kappa..function..kappa..kappa..kappa. ##EQU00001##
Here, the dimensionless flow rate a calculated by the dimensionless
flow rate calculation section 14 is defined as shown by the
following expression (2).
.sigma..kappa..function..kappa..kappa..kappa. ##EQU00002##
The amount of intake air Qa can be represented by the following
expression (3) by assigning expression (2) to expression (1).
Qa=a.sub.0CA.sub.t.sigma. (3)
Here, note that the sound speed a.sub.0 in the atmosphere is
represented by the following expression (4) by using a gas constant
R and the intake air temperature To.
.kappa..times..times. ##EQU00003##
In addition, upon transformation of expression (3), the effective
opening area CAt represented by the product of the flow coefficient
C and the opening area At of the throttle valve 4 can be calculated
by the following expression (5) when the target amount of intake
air Qa* required to achieve the target torque, the speed of sound
a.sub.0 in the atmosphere and the dimensionless flow rate .sigma.
are provided.
.sigma. ##EQU00004##
Accordingly, the target effective opening area calculation section
11 in the ECU 9 calculates the target effective opening area CAt*
to achieve the target amount of intake air Qa* by using expression
(5) based on the target amount of intake air Qa*, the speed of
sound a.sub.0 in the atmosphere and the dimensionless flow rate
.sigma..
Thus, by calculating the target effective opening area CAt* based
on the volumetric flow formula of the throttle type flow meter
represented by expression (1), the target effective opening area
CAt* to adequately achieve the target amount of intake air Qa* can
be calculated even if the operating state of the engine 1 is
changed resulting from a change of the environmental condition, the
introduction of EGR (opening of the EGR valve 8), etc.
However, the calculation of the speed of sound a.sub.0 in the
atmosphere, which is required for the calculation of the target
effective opening area CAt*, by using expression (4) in the ECU 9
becomes huge in the calculation load and hence not practicable.
Accordingly, in order to suppress or reduce the calculation load in
the ECU 9, the sound speed calculation section 12 calculates the
theoretical value of the speed of sound a.sub.0 in the atmosphere
in advance, stores it as map data with respect to the intake air
temperature To, and calculates the speed of sound a.sub.0 in the
atmosphere by using the intake air temperature To before
calculation processing in the target effective opening area
calculation section 11.
Similarly, the calculation of the dimensionless flow rate .sigma.
required for the calculation of the target effective opening area
CAt* in the ECU 9 by using expression (2) becomes huge in the
calculation load and hence is not practicable.
Accordingly, in order to suppress or reduce the calculation load in
the ECU 9, the dimensionless flow rate calculation section 14
calculates the theoretical value of the dimensionless flow rate a
in advance, stores it as map data with respect to the pressure
ratio of the intake manifold pressure Pe and the atmospheric
pressure Po, and calculates the dimensionless flow rate .sigma. by
using the pressure ratio Pe/Po of the intake manifold pressure Pe
and the atmospheric pressure Po calculated by the pressure ratio
calculation section 13 before calculation processing in the target
effective opening area calculation section 11.
However, it is generally known that when the pressure ratio Pe/Po
is equal to or less than a first predetermined value (about 0.528
in case of air), the flow rate of air passing through the throttle
valve 4 is saturated in the following cases (so-called choking). In
addition, it is also known that when such a choke occurs, the
dimensionless flow rate .sigma. calculated by expression (2)
becomes a definite or fixed value.
Accordingly, the pressure ratio calculation section 13 includes a
pressure ratio fixing section (not shown) which can deal with the
occurrence of choking by fixedly setting the pressure ratio Pe/Po
to the first predetermined value when the pressure ratio Pe/Po is
equal to or less than the first predetermined value.
Here, note that instead of fixedly setting the pressure ratio Pe/Po
to the first predetermined value in the pressure ratio calculation
section 13, the map value of the dimensionless flow rate .sigma.
corresponding to the pressure ratio Pe/Po in the dimensionless flow
rate calculation section 14 may be set to the same value as in the
case of the first predetermined value, in a region in which the
pressure ratio Pe/Po is equal to or less than the first
predetermined value.
On the other hand, when the pressure ratio Pe/Po becomes equal to
or larger than a certain value, the air flow sensor 2 and the
intake manifold pressure sensor 7 are subjected to the influence of
the pulsation of intake air, so there is a possibility that an
error might occur in the measured value of the amount of intake air
Qa with respect to the actual amount of intake air. Besides, there
is also a possibility that the calculation of the dimensionless
flow rate .sigma. might be subjected to the great influence of a
measurement error of the intake manifold pressure Pe due to the
intake air pulsation.
Accordingly, when the pressure ratio Pe/Po is larger than the first
predetermined value, the pressure ratio fixing section (not shown)
in the pressure ratio calculation section 13 suppresses the
influence of the intake air pulsation thereby to ensure the
controllability of the throttle valve 4 by dealing with the
pressure ratio Pe/Po as the first predetermined value.
Here, note that instead of fixedly setting the pressure ratio Pe/Po
to the first predetermined value in the pressure ratio calculation
section 13, the map value of the dimensionless flow rate .sigma.
corresponding to the pressure ratio Pe/Po in the dimensionless flow
rate calculation section 14 may be set to the same value as in the
case of the first predetermined value, in a region in which the
pressure ratio Pe/Po is equal to or larger than the first
predetermined value.
Hereinafter, the target opening calculation section 15 calculates
the target opening TP* by using the target effective opening area
CAt* calculated by the target effective opening area calculation
section 11.
At this time, the target opening calculation section 15 obtains in
advance the relation between the measured value of the throttle
opening TP and the effective opening area CAt calculated from the
measured value of the amount of intake air Qa according to
expression (5), stores it as a two dimensional map in which the
throttle opening TP and the effective opening area CAt correspond
to each other one by one, and calculates the target opening TP*
corresponding to the target effective opening area CAt* by using
this two dimensional map.
As a result, the two dimension map of the throttle opening TP and
the effective opening area CAt can be easily prepared, thus making
it possible to reduce the man-hours for setting to a substantial
extent.
Subsequently, upon controlling the throttle valve 4 so as to attain
the target opening TP* calculated by the target opening calculation
section 15, the throttle opening control section in the processing
unit 9b calculates the throttle opening learning value so as to
decrease an error between the target amount of intake air Qa* and
the actual amount of intake air Qa resulting from the variations of
the throttle body and the various kinds of sensors 30, various
estimation errors, etc.
Now, specific reference will be made to calculation processing for
a throttle opening learning value TPLRN according to the first
embodiment of the present invention while referring to FIG. 4 and
FIG. 5.
FIG. 4 is an explanatory view that schematically illustrates the
calculation processing for the throttle opening learning value
TPLRN, and FIG. 5 is a functional block diagram that schematically
illustrates a peripheral construction of a throttle opening
learning value calculation processing section 22 in the throttle
opening control section 17.
In FIG. 5, the throttle opening control section 17 in the
processing unit 9b of the ECU 9 is provided with an actual
effective opening area calculation section 18, a learning throttle
opening calculation section (hereinafter referred to as a "learning
opening calculation section") 19, the learning basic value
calculation section 20 connected to the target opening calculation
section 15, a post-correction integration processing section 21
that integrates a learning basic value .DELTA.TP, the throttle
opening learning value calculation section 22, and the learning
corrected target throttle opening calculation section 23
(hereinafter referred to as a "learning corrected target opening
calculation section").
Here, note that the configuration upstream of the target opening
calculation section 15 is similar to that in the above-mentioned
(see FIG. 3) and hence is omitted.
The actual effective opening area calculation section 18 calculates
an actual effective opening area CAtr of the throttle valve 4
according to the throttle opening control section 17 based on the
actual amount of intake air Qa when the throttle valve 4 is
controlled to the target opening TP*.
At this time, the actual effective opening area calculation section
18 calculates the actual effective opening area CAtr of the
throttle opening control section 17, as shown by the
above-mentioned expression (5), by applying the amount of intake
air Qa, the atmospheric pressure Po, the intake manifold pressure
Pe, and the intake air temperature To the flow rate formula of a
so-called throttle type flow meter, and inputs it to the learning
throttle opening calculation section 19.
The learning throttle opening calculation section 19 calculates a
learning throttle opening TPi (hereinafter referred to as a
"learning opening") from the actual effective opening area CAtr by
using a correspondence or correlation map between the throttle
opening TP and the effective opening area CAt that are suited to
each other in advance (hereinafter referred to as a "CAt-TP map"),
and inputs it to the learning basic value calculation section
20.
The learning basic value calculation section 20 calculates a
deviation .DELTA.TP (=TP*-TPi) between the target opening TP* and
the learning opening TPi as the learning basic value, and inputs it
to the post-correction integration processing section 21.
The post-correction integration processing section 21 integrates
the value obtained by multiplying the learning basic value
.DELTA.TP by a correction factor Kc (0.ltoreq.Kc.ltoreq.1) in a
sequential manner (or by applying filtering processing to the
learning basic value .DELTA.TP), and inputs a value, which is
obtained by removing an instantaneous variation from the learning
basic value .DELTA.TP, to the throttle opening learning value
calculation section 22.
The throttle opening learning value calculation section 22
calculates the throttle opening learning value TPLRN based on the
learning basic value .DELTA.TP obtained through the
post-integration processing section 21, and inputs it to the
learning corrected target opening calculation section 23.
The learning corrected target opening calculation section 23 adds
the throttle opening learning value TPLRN and the target opening
TP* calculated by the target opening calculation section 15 to each
other thereby to calculate a learning corrected target throttle
opening (hereinafter referred to as a "learning corrected target
opening") TPLRN*.
Thus, the throttle opening control section 17 calculates the
throttle opening learning value TPLRN based on the learning basic
value .DELTA.TP (deviation between the target opening TP* and the
learning opening TPi), and controls the throttle opening TP by
using a learning corrected target opening TPLRN* that is obtained
by correcting the target opening TP* by the throttle opening
learning value TPLRN.
Hereinafter, specific reference will be made to the learning
function of the throttle opening control section 17 while referring
to FIG. 4 together with FIG. 3 and FIG. 5.
Here, assuming that the effective opening area CAt and the throttle
opening TP correspond to each other one by one, when there exists
an error between the target amount of intake air Qa* and the actual
amount of intake air Qa, there will also be an error between the
target effective opening area CAt* calculated from the target
amount of intake air Qa* by the target effective opening area
calculation section 11 and the actual effective opening area CAtr
that is obtained by the actual effective opening area calculation
section 18 by applying the amount of intake air Qa to expression
(5).
For example, let us consider a case where as shown in FIG. 4, there
exists an error between the CAt-TP map for control (see a broken
line) and an actual relation between the effective opening area CAt
and the throttle opening TP (hereinafter referred to as an "actual
CAt-TP relation") (see a solid line) that are estimated while
including the variation of the throttle body of the engine 1 to be
currently controlled and the variations of the various kinds of
sensors 30 that serve to measure the intake manifold pressure Pe,
the atmospheric pressure Po, the intake air temperature To and so
on.
First of all, the target effective opening area calculation section
11 calculates the target effective opening area CAt* from the
target amount of intake air Qa*, as previously stated, and the
target opening calculation section 15 calculates the target opening
TP* from the presuited CAt-TP map (see a broken line in FIG. 4) by
using the target effective opening area CAt*.
The relation between the target effective opening area CAt* and the
target opening TP* calculated at this time is indicated at point a
on the CAt-TP map in FIG. 4.
However, when an error exists between the CAt-TP map (broken line)
and the actual CAt-TP relation (solid line), as shown in FIG. 4,
the actual effective opening area CAtr at point b on the actual
CAt-TP relation (solid line) corresponding to the target opening
TP* is different from the target effective opening area CAt*, so
the actual amount of intake air Qa that is obtained when the
throttle opening TP is controlled to the target opening TP* will
not coincide with the target amount of intake air Qa*.
Accordingly, in order to calculate a learning value for correction
of this error, the actual effective opening area calculation
section 18 first calculates the actual effective opening area CAtr
based on the actual amount of intake air Qa measured at the time
when the throttle valve 4 is controlled to the target opening
TP*.
The relation between the actual effective opening area CAtr
calculated by the actual effective opening area calculation section
18 and the target opening TP* is shown at point b on a curve of the
actual CAt-TP relation (solid line) in FIG. 4.
In FIG. 4, to achieve the target effective opening area CAt* (the
target amount of intake air Qa*), it is necessary to control the
throttle opening TP to point d on the curve of the actual CAt-TP
relation (solid line), it is necessary to calculate a difference
between the point a and the point d as a learning value.
At this time, assuming that the CAt-TP map (broken line) and the
actual CAt-TP relation (solid line) are locally substantially in a
parallel relation with respect to each other, as shown in FIG. 4,
the learning opening calculation section 19 calculates the learning
opening TPi by using the CAt-TP map (broken line) based on the
actual effective opening area CAtr calculated from the amount of
intake air Qa obtained when the throttle valve 4 is controlled to
the target opening TP*.
The relation between the learning opening TPi thus calculated and
the actual effective opening area CAtr is indicated at point c on
the CAt-TP map in FIG. 4.
Accordingly, the learning basic value calculation section 22
calculates the learning basic value .DELTA.TP on the assumption
that the throttle opening deviation .DELTA.TP (=TP*-TPi) between
the target opening TP* and the learning opening TPi indicated by a
difference between point b and point c is substantially equal to
the learning basic value between point a and point d.
The learning basic value .DELTA.TP calculated by the learning basic
value calculation section 22 includes an instantaneous variation,
so the post-correction integration processing section 21
sequentially integrates the value obtained by multiplying the
learning basic value .DELTA.TP by the correction factor Kc (or by
applying filtering processing), and then, the throttle opening
learning value calculation section 22 calculates the throttle
opening learning value TPLRN.
Finally, the learning corrected target opening calculation section
23 adds the throttle opening learning value TPLRN to the target
opening TP* to calculate the learning corrected target opening
TPLRN*.
Hereinafter, by controlling the throttle opening TP with the use of
the learning corrected target opening TPLRN*, the throttle opening
control section 17 decreases an error or difference between the
target amount of intake air Qa* and the amount of intake air
Qa.
Accordingly, it is possible to learn and correct the relation
between the effective opening area CAt and the throttle opening TP
in such a manner that upon calculation of the throttle opening TP
for obtaining the target amount of intake air Qa*, the target
amount of intake air Qa* can be adequately attained with respect to
variations in the throttle body and various sensors or various
kinds of estimation errors.
At this time, if an error or difference between the CAt-TP map
(broken line) and the actual CAt-TP relation (solid line) is in a
substantially constant or fixed (substantially parallel) relation,
it is possible to adequately control the throttle valve 4 in the
entire operating range of the engine 1 even when the throttle
opening learning value TPLRN is used independently for feedback
control.
Embodiment 2
Though not particularly referred to in the above-mentioned first
embodiment, for example as shown in FIG. 6, in case where the
CAt-TP map (see a broken line) crosses with respect to the actual
CAt-TP relation (see a solid line), or an error of the CAt-TP map
(see an alternate long and short dash line) with respect to the
actual CAt-TP relation is not constant (parallel), there is a
possibility that problems such as a follow-up delay, an overshoot,
etc., might occur at the time of a transient operation if the
throttle opening learning value TPLRN is used independently.
Accordingly, in order to deal with the case where the CAt-TP maps
(see the broken line and the alternate long and short dash line)
are not constant with respect to the actual CAt-TP relation (solid
line) (see FIG. 6), it is desirable that as shown in FIG. 7,
provision be made for a throttle opening learning value
distribution section 24 that is arranged downstream of the throttle
opening learning value calculation section 22 for distributing the
throttle opening learning value TPLRN to a real-time learning value
TPR to be used for feedback control and a long time learning value
TPL to be stored in individual learning regions corresponding to
portions of a CAt axis (the axis of abscissa in FIG. 4 and FIG. 6)
of the CAt-TP map.
As a result, the sum of a value on the CAt-TP map and the long time
learning value TPL can be brought close to the actual CAt-TP
relation. In addition, an instantaneous error can be absorbed by
the feedback control together with the use of the real-time
learning value TPR.
Hereinafter, specific reference will be made to a second embodiment
of the present invention while referring to explanatory views in
FIG. 8 and FIG. 9 together with a functional block diagram in FIG.
7.
In FIG. 7, a throttle opening control section 17A according to the
second embodiment of the present invention includes the throttle
opening learning value distribution section 24 connected to the
throttle opening learning value calculation section 22, a real-time
learning value calculation section 25 connected to the throttle
opening learning value distribution section 24 through a switching
section 24a, a long time learning value calculation section 26
connected to the throttle opening learning value distribution
section 24 through a switching section 24b, a monotonous increase
processing section 27 connected to the long time learning value
calculation section 26, a long time learning value storage section
28 connected to the monotonous increase processing section 27, and
a correction throttle opening learning value calculation section 29
(hereinafter referred to as a "correction opening learning value
calculation section") connected to the real-time learning value
calculation section 25 and the long time learning value storage
section 28.
Here, note that the configuration upstream of the throttle opening
learning value calculation section 22 is similar to that in the
above-mentioned (see FIG. 5) and hence is omitted.
In this case, the throttle opening learning value TPLRN is
distributed to and stored in at least one of the real-time learning
value TPR being updated in real time, and the long time learning
value TPL corresponding to the individual learning regions
according to the effective opening area axis (CAt axis) of the
CAt-TP map.
Also, the long time learning value TPL is stored in at least one of
a learning region corresponding to the target effective opening
area CAt* and a learning region corresponding to the actual
effective opening area CAtr.
First of all, the throttle opening learning value distribution
section 24 distributes the throttle opening learning value TPLRN to
the real-time learning value TPR and the long time learning value
TPL at a predetermined ratio.
The switching section 24a inputs "0" to the real-time learning
value calculation section 25 when a predetermined reset condition
holds, and inputs the last real-time learning value TPR(n-1) to the
real-time learning value calculation section 25 when a
predetermined update inhibiting condition holds, whereas the
switching section 24a inputs the current throttle opening learning
value TPLRN to the real-time learning value calculation section 25
when the reset condition and the update inhibiting condition of the
real-time learning value TPR do not hold.
Accordingly, when the reset condition and the update inhibiting
condition (to be described later) of the real-time learning value
TPR do not hold, the real-time learning value calculation section
25 calculates a final real-time learning value TPR based on the
throttle opening learning value TPLRN.
The switching section 24b inputs the last long time learning value
TPL(n-1) to the long time learning value calculation section 26
when the predetermined update inhibiting condition holds, and
inputs the current throttle opening learning value TPLRN to the
long time learning value calculation section 26 when the update
inhibiting condition of the long time learning value TPL does not
hold.
Accordingly, when the update inhibiting condition of the long time
learning value TPL do not hold, the final long time learning value
calculation section 26 calculates a final long time learning value
TPL for each of the learning regions according to the portions of
the CAt axis of the CAt-TP map based on the throttle opening
learning value TPLRN.
Here, note that as a concrete example of the update inhibiting
condition in the switching sections 24a, 24b, the updates of the
real-time learning value TPR and the long time learning value TPL
may be inhibited when the pressure ratio Pe/Po of the intake
manifold pressure Pe (the intake pipe internal pressure) to the
atmospheric pressure Po indicates a value equal to or larger than a
first predetermined value.
In addition, as a concrete example of the reset condition in the
switching section 24a, the real-time learning value TPR may be
reset in a period in which the time elapsed after the time change
rate dQa*/dt of the target amount of intake air Qa* has reached a
second predetermined value or more indicates a value less than or
equal to a third predetermined value. This condition is at the same
time used as the update inhibiting condition of the long time
learning value TPL, too.
The monotonous increase processing section 27 limits the long time
learning value TPL in such a manner that the CAt-TP map and the
actual CAt-TP relation (the relation between the effective opening
area CAt and the throttle opening TP of the throttle opening
control section 17A) after corrected by addition thereto of the
long time learning value TPL become monotonously increasing.
The long time learning value storage section 28 stores the long
time learning value TPL through the monotonous increase processing
section 27.
The correction opening learning value calculation section 29 is in
the form of an adding section that serves to add the real-time
learning value TPR and the long time learning value TPL to each
other, and inputs the result of the addition to the learning
corrected target opening calculation section 23 as a correction
throttle opening learning value TPLRNi (hereinafter referred to as
a "correction opening learning value").
The long time learning value storage section 28 in the throttle
opening control section 17A functions as a backup memory. That is,
when the engine 1 is stopped or when the power supply for the
control apparatus for an internal combustion engine is turned off,
the real-time learning value TPR is reset, and the long time
learning value TPL is held in the long time learning value storage
section 28 (backup memory).
In addition, the update of the real-time learning value TPR is
inhibited in a period of time in which the time elapsed after the
starting of the engine 1 indicates a value within a fourth
predetermined value, and the update of the long time learning value
TPL is inhibited in a period of time in which the time elapsed
after the starting of the engine 1 indicates a value that is equal
to or larger than the fourth predetermined value and within a fifth
predetermined value.
Further, when the number of revolutions per minute of the engine 1
indicates a value less than or equal to a sixth predetermined value
that is lower than a target number of revolutions per minute of the
engine 1 during idling, the update of the long time learning value
TPL is inhibited.
Next, specific reference will be made to the calculation processing
of the long time learning value TPL in each learning region
according to the second embodiment of the present invention, as
illustrated in FIG. 7, while referring to FIGS. 8 and 9 together
with FIG. 4.
FIGS. 8 and 9 are explanatory views that schematically illustrate
storage processing and monotonously increasing processing,
respectively, for the long time learning value according to the
second embodiment of the present invention.
In FIG. 4, as previously stated, the throttle opening learning
value TPLRN calculates a difference .DELTA.TP between point b and
point c (a throttle opening deviation between the target opening TP
and the learning opening TPi) as a learning basic value, and
applies the learning basic value .DELTA.TP as a learning value
between point a and point d.
Here, let us consider the case where the throttle opening learning
value TPLRN is distributed to and stored in learning regions
corresponding to a one to one distribution for example with respect
to the CAt axis of the CAt-TP map.
At this time, as shown in FIG. 8, the long time learning value TPL
can be stored in at least one of a learning region corresponding to
the CAt axis before and after the target effective opening area
CAt* and a learning region corresponding to the CAt axis before and
after the actual effective opening area CAtr.
Here, note that the long time learning value TPL stored in a
learning region corresponding to each CAt axis is calculated by
adding a predetermined value to the last long time learning value
TPL(n-1), or by adding a value corresponding to a ratio of
distances between the CAt axes before and after the target
effective opening area CAt* and the actual effective opening area
CAtr to the last long time learning value TPL(n-1).
In addition, if the long time learning value TPL is stored with
both the target effective opening area CAt* and the actual
effective opening area CAtr, the convergence time of the long time
learning value TPL can be shortened.
In case where the long time learning value TPL is calculated in
this manner, the learnable condition is only the case where the
update inhibiting condition does not hold (to be described later),
so the case of actual learning being carried out is limited only to
a region in which steady-state operation is regularly used.
In general, the throttle opening TP and the amount of intake air Qa
are in a monotonously increasing relation, so the effective opening
area CAt and the throttle opening TP should be in a monotonously
increasing relation.
However, when learning is locally carried out, as shown by a broken
line and a broken line frame in FIG. 9, the sum of the value of the
CAt-TP map (see a solid line) and the long time learning value TPL
(see a broken line) might not become monotonously increasing.
In this case, the learning corrected target opening TPLRN*
decreases in spite of the increasing target amount of intake air
Qa* for example, so there arise problems such as reduction in the
output power of the engine 1, the mislearning of the throttle
opening learning value TPLRN, etc.
Accordingly, the monotonous increase processing section 27 performs
the processing of adding a predetermined value to the long time
learning value TPL thereby to limit the long time learning value
TPL in such a manner that the sum of the value of the CAt-TP map
(solid line) and the long time learning value TPL (see a dotted
line) becomes monotonously increasing, as shown by a dotted line
and a dotted line frame in FIG. 9. As a result, the mislearning and
malfunction of the throttle opening learning value TPLRN can be
prevented.
Hereinafter, specific reference will be made to the monotonously
increasing processing according to the monotonous increase
processing section 27.
First of all, by using a CAt axis number n, the long time learning
value currently being learned is set as TPL(n), and the range that
can be taken by the CAt axis number n currently being learned is
set to "1.ltoreq.n.ltoreq.CAt axis number".
Here, the long time learning value TPL after the monotonously
increasing correction can be calculated by repeating the
calculation of the following expression (6) for a long time
learning value TPL (m+1+i) that is in a region in which the CAt
axis number n thereof is larger than a predetermined value m.
TPL(m+1+i) =max{CAt map value(m+i)+TPL(m+i)+predetermined value,
CAt map value(m+1+i)+TPL(m+1+i)}-CAt map value(m+1+i) (6) where
variable i sequentially increases from "0" up to "CAt axis
number-(m+1)" at the time of repeating the calculation.
On the other hand, a long time learning value TPL (m-1-i), which is
in a region where the CAt axis number n thereof is less than the
predetermined value m, can be calculated by repeating the
calculation of the following expression (7). TPL(m-1-j) =min{CAt
map value(m-j)+TPL(m-j)-predetermined value, CAt map
value(n-1-j)+TPL(m-1-j)}-CAt map value(m-1-j) (7) where a variable
j sequentially increases from "0" up to "m-2" at the time of
repeating the calculation.
After execution of the calculations of the above-mentioned
expressions (6), (7), the long time learning value storage section
28 stores a final long time learning value TPL in each learning
region.
As shown in FIG. 7, in case where the throttle opening learning
value TPLRN is distributed to the real-time learning value TPR and
the long time learning value TPL which is then stored, the
correction opening learning value calculation section 29 adds the
real-time learning value TPR and the long time learning value TPL
corresponding to an engine operating range to each other thereby to
calculate a correction opening learning value TPLRNi, and inputs it
to the learning corrected target opening calculation section
23.
Accordingly, the learning corrected target opening calculation
section 23 calculates the learning corrected target opening TPLRN*
(=TPLRNi+TP*) by using the correction opening learning value TPLRNi
in place of the throttle opening learning value TPLRN.
As described above, the calculation of the throttle opening
learning value TPLRN is performed, and at the same time, the
calculation and storing of the long time learning value TPL based
on the throttle opening learning value TPLRN are also carried out,
but such learning processing can not be performed in all the
operating ranges, so learning inhibiting processing is needed.
Hereinafter, specific reference will be made to a learning
inhibiting condition according to the second embodiment of the
present invention.
As stated above, the air flow sensor 2 is subjected to the
pulsation of intake air when the pressure ratio Pe/Po of the intake
manifold pressure Pe to the atmospheric pressure Po increases to a
certain extent. As a result, an error might occur between an actual
amount of intake air and a measured amount of intake air, so in
such an operating range, the throttle opening learning value TPLRN
can not be calculated accurately.
Accordingly, when the pressure ratio Pe/Po indicate a value equal
to or larger than the above-mentioned first predetermined value,
the switching sections 24a, 24b select the last real-time learning
value TPR(n-1) and the last long time learning value TPL(n-1),
respectively, and inhibit the updates of the real-time learning
value TPR and the long time learning value TPL. As a result, the
mislearning of the throttle opening TP due to the influence of the
intake air pulsation can be prevented.
In addition, when the target amount of intake air Qa* is suddenly
changed during the transient operation or the like, a certain time
will be needed until the time when the amount of intake air Qa
responds to the change of the target amount of intake air Qa*, due
to the calculation time delay until the completion of calculation
of the target opening TP*, the response delay until the throttle
opening TP arrives at the target opening TP*, the response delay
until the flow speed near the air flow sensor 2 is changed due to
the change of the throttle opening, the response delay of the air
flow sensor 2 itself, and so on.
Accordingly, when the time elapsed after the rate of change of the
target amount of intake air Qa* became the second predetermined
value or above indicates a value within the third predetermined
value, the switching section 24b inhibits the update of the long
time learning value TPL. As a result, the mislearning of the long
time learning value TPL due to a response delay in the amount of
intake air Qa can be prevented.
Here, note that in the condition at this time, if the update of the
real-time learning value TPR is also inhibited by the switching
section 24a, when it is considered that the value of the CAt-TP map
(broken line) in FIG. 6 changes across a crossing point thereof
with the actual CAt-TP relation (solid line) for example, the sign
of the real-time learning value TPR is reversed before and after
the crossing point, so there arise problems such as the occurrence
of overshooting, an increased time of convergence to the target
amount of intake air Qa*, etc.
Accordingly, when the time elapsed after the rate of change of the
target amount of intake air Qa* became the second predetermined
value or above indicates a value within the third predetermined
value, the switching section 24a performs reset processing on the
real-time learning value TPR (TPR=0). Thus, overshooting can be
suppressed, whereby it is possible to prevent an increase in the
convergence time to the target amount of intake air Qa*.
Further, when the engine 1 is stopped or the power supply of the
ECU 9 is turned off, the learning processing of the throttle
opening TP can not be executed, so the switching section 24a
performs the reset processing of the real-time learning value
TPR.
On the other hand, by holding the long time learning value TPL in
the long time learning value storage section 28 (backup memory), it
is possible to adequately control the throttle valve 4 so as to
achieve the target amount of intake air Qa* even at the next
restart of the engine 1.
Here, note that at the start of the engine 1, the air near the air
flow sensor 2 does not generally move during the time the engine 1
is consuming the air in the surge tank 6, so a certain period of
time is required until the air near the air flow sensor 2 begins to
flow after the start of the engine 1 so as to enable the amount of
intake air Qa to be measured in an accurate manner.
In view of this, when the time elapsed after the starting of the
engine 1 indicates a value within the fourth predetermined value,
the storage section 24 inhibits the update of the real-time
learning value TPR, whereby an error in the calculation of the
real-time learning value TPR due to the influence of the amount of
intake air Qa can be prevented.
In addition, until the time the number of revolutions per minute of
the engine 1 is converged into the one at the time of idling after
the starting of the engine 1, variations in the number of
revolutions per minute of the engine 1 and in the amount of intake
air Qa are intense, so it is undesirable to store the throttle
opening learning value TPLRN as the long time learning value
TPL.
Accordingly, when the time elapsed after the starting of the engine
1 indicates a value within the fifth predetermined value
(.gtoreq.fourth predetermined value), the switching section 24b
inhibits the update of the long time learning value TPL, whereby
the mislearning of the long time learning value TPL can be
prevented.
In this case, when the time elapsed after the starting of the
engine 1 is between the fourth predetermined value and the fifth
predetermined value, the real-time learning value TPR is updated,
but the real-time learning value TPR acts as feedback control.
Thus, by updating the learning value, the throttle opening TP is
controlled so as to achieve the target amount of intake air Qa*,
whereby an engine stall due to the reduction in the number of
engine revolutions per minute, which might otherwise be caused
after the starting of the engine 1, can be prevented.
Moreover, in cases where the number of revolutions per minute of
the engine 1 falls greatly below that during idling immediately
before the engine 1 is stopped or in accordance with load
variation, etc., variations in the number of engine revolutions per
minute and in the amount of intake air Qa are intense, too, so it
is undesirable to store the throttle opening learning value TPLRN
as the long time learning value TPL.
Accordingly, when the number of revolutions per minute of the
engine 1 falls below the sixth predetermined value which is smaller
than the number of revolutions per minute of the engine 1 during
idling, the switching section 24b also inhibits the update of the
long time learning value TPL.
On the other hand, the real-time learning value TPL acts as
feedback control, so the engine stall can be prevented by updating
the learning value thereby to control the throttle opening so as to
achieve the target amount of intake air Qa*.
While the invention has been described in terms of preferred
embodiments, those skilled in the art will recognize that the
invention can be practiced with modifications within the spirit and
scope of the appended claims.
* * * * *